The Department of Transportation's Federal Highway Commission has released the latest report on Traffic Volume Trends, data through April. Travel on all roads and streets changed by 0.9% (2.3 billion vehicle miles) for May 2013 as compared with May 2012. Travel for the month is estimated to be 262.1 billion vehicle miles. Cumulative Travel for 2013 changed by -0.5%. Cumulative estimate for the year is 1,203.6 billion vehicle miles of travel. (PDF report). Both the civilian population-adjusted data (age 16-and-over) and total population-adjusted data are hovering just off their post-financial crisis lows.
Here is a chart that illustrates this data series from its inception in 1970. I'm plotting the "Moving 12-Month Total on ALL Roads," as the DOT terms it. See Figure 1 in the PDF report, which charts the data from 1987. My start date is 1971 because I'm incorporating all the available data from earlier DOT spreadsheets.
The rolling 12-month miles driven contracted from its all-time high for 39 months during the stagflation of the late 1970s to early 1980s, a double-dip recession era. The most recent decline has lasted for 65 months and counting — a new record, but the trough to date was in November 2011, 48 months from the all-time high.
The Population-Adjusted Reality
Total Miles Driven, however, is one of those metrics that should be adjusted for population growth to provide the most revealing analysis, especially if we're trying to understand the historical context. We can do a quick adjustment of the data using an appropriate population group as the deflator. I use the Bureau of Labor Statistics' Civilian Noninstitutional Population Age 16 and Over (FRED series CNP16OV). The next chart incorporates that adjustment with the growth shown on the vertical axis as the percent change from 1971.
Clearly, when we adjust for population growth, the Miles-Driven metric takes on a much darker look. The nominal 39-month dip that began in May 1979 grows to 61 months, slightly more than five years. The trough was a 6% decline from the previous peak.
The population-adjusted all-time high dates from June 2005. That's 93 months — over seven years ago. The latest data is 8.91% below the 2005 peak, fractionally off the post-Financial Crisis low of -8.93% set two months earlier. Our adjusted miles driven based on the 16-and-older age cohort is about where we were as a nation in January of 1995.
Here is a closer look at the series since 2000. We can see that the decline leveled off in 2010, rolled over in 2011, leveled off in the first half of 2012 and then rolled over again.
About that Population Adjustment...
I've frequently been asked why I use the CNP16OV data for the population adjustment, often with the suggestion that it would make more sense to limit the population to licensed drivers. For openers, I don't know of a valid source for the driver-licensed population. Moreover, the correlation between license holders and actual drivers is not a reliable one. Many license holders in households do not drive, especially in their older years. According to Census Bureau data on gasoline sales (courtesy of Harry Dent's research on demand curves), dollars spent on gasoline peaks for people in their late 40s and falls off rather quickly after that.
In fact, I think there's a good case for using the Census Bureau's mid-month estimates of total population (POPTHM) rather than civilians age 16 and over for the population adjustment. The reason is that a portion of total miles driven is specifically to support children's needs (day care, schools, children's activities, etc.) and the needs of elders who might have licenses but no longer drive. Ultimately the division of miles driven by either population group (CNP16OV or POPTHM), while not a perfect match with drivers, is a consistent and relevant metric for evaluating economic growth.
Here is the same population-adjusted chart, this time with the total population for the adjustment. In the total-population adjusted version the latest data point of -7.45%, fractionally off the post-recession low of -7.50% set two months earlier.
What about the impact of volatile gasoline prices? How much of a factor has that been in the trend? I'll close with an overlay of the population-adjusted miles driven and gasoline prices since the early 1990s.
As is readily apparent, the correlation is fairly weak over the entire timeframe (+0.30). And, despite the volatility in gasoline prices since the onset of the Great Recession, the correlation since December 2007 has been even weaker (-0.26). There are profound behavioral issues apart from gasoline prices that are influencing miles driven. These would include the demographics of an aging population in which older people drive less, continuing high unemployment, and the ever-growing ability to work remote in the era of the Internet.
See the latest study by the Frontier Group, A New Direction: Our Changing Relationship with Driving and the Implications for America's Future (PFD format).
The DOT "miles driven" metric is also interesting to study in the context of gasoline volume sales, which I also update monthly: Gasoline Volume Sales, Demographics and our Changing Culture